View System Configuration & Spec Sheet
talha@infer-workstation
-----------------------
Host: Custom Workstation (Fedora Linux / Kernel 6.x)
Role: AI Systems & Inference Engineer
Focus: Local LLM Optimization & Quantization (GGUF/AWQ)
Compute: Vulkan ICD, DirectML, llama.cpp Compute Layers
Engines: Python 3.12, C++20, Rust, FastAPI, PyTorch
Infra: Docker, PostGIS, pgvector, Linux Kernel Internals
Status: Available for high-impact contracts & remote roles| Engine / Platform | Architecture & Scope | Core Stack | Status |
|---|---|---|---|
| VulkanIlm | High-throughput local LLM inference wrapper using Vulkan backends via llama.cpp for non-CUDA/consumer GPUs. Up to 33× speedup over CPU execution. |
Python C++ Vulkan GGUF |
|
| Buildables AI Track | Production LLM curriculum & execution code: Advanced hybrid RAG, structured tool-calling, and evaluation suites. | FastAPI pgvector LangGraph |
|
| Urban-Intel Engine | Spatial intelligence platform integrating Vision Transformers, automated raster ingestion, and vector geometry. | FastAPI PostGIS PyTorch |
|
Synced via local workstation session |
|




